Introduction

Mafic magma has been recognized to play an important role in the process of granite melt generation. The processes of felsic and mafic magma interactions can be observed at different scales: from outcrop scales to micro-chemical and isotopic features. The scale of the observation depends on the range of the process involved: from mixing—where different magmas’ identity is not apparent—to mingling, where the individual components from different magmas can be recognized (Menéndez and Ortega 1999, Didier and Barbarin 1991 and references therein). Fundamental features that unequivocally prove the presence of mixing—mingling processes are textural ones.

According to Hibbard (1991), no single texture can be used to prove undoubtedly the presence of hybridization in the granitic rocks, so the term “textural assemblage” was proposed to define the combination of textures, forming together a good proof of mixing—mingling phenomena. Those are as follows: rapakivi and/or antirapakivi texture (Hibbard 1991, Nekvasil 1991), poikilitic crystals of quartz and K-feldspars, sphene-rich ocellar texture (Hibbard 1991, Baxter and Feely 2002), blade “hydrogenic” biotite (Hibbard 1991; Vernon 2004), zoned K-feldspar phenocrysts, with rows of plagioclase, biotite, hornblende etc. inclusions (Vernon 2004, Hibbard 1991; Vernon and Paterson 2008, Słaby et al. 2007a, Słaby et al. 2007b), spike zones in plagioclases and boxy cellular morphology of plagioclase crystals, acicular apatite and mixed apatite morphologies (Hibbard 1991, Baxter and Feely 2002) and small plagioclase laths (Hibbard 1991).

The Western Tatra granitoids (common Tatra type) were formerly interpreted as purely S-type rocks, formed during a prolonged 360–350 Ma magmatic episode (Kohút and Janak 1994; Poller et al. 2001a). In this paper, we discuss the age as well as textural and mineralogical evidences of magma hybridization in the Western Tatra granitoids. The importance of mafic and felsic magma mixing for the formation of the Western Tatra granitoid pluton and the connection to quartz-diorite mafic precursors are also discussed.

Geological setting

The Tatra Mountains are one of the so-called core mountains in the Central Western Carpathians (Fig. 1a), which are uplifted portions of the Variscan crust tectonically emplaced among the Alpine structures of the Carpathian mountain chain. The Tatra Mountains consist of a Variscan crystalline core, overlain by Mesozoic sedimentary formations. The crystalline core comprises composite granitoid intrusion and its metamorphic envelope (Fig. 1b). In the granitoid body four petrographic types of granitoids can be distinguished (Morozewicz 1914; Kohút and Janak 1994). The common Tatra granodiorite-tonalite (S-type) is volumetrically predominant tongue-shaped intrusion, dated at 360–350 Ma (Fig. 1b; Poller et al. 2000, Poller et al. 2001a, b). Quartz-diorites (I-type mingled hybrid, interpreted as magmatic precursors) are present as sills inside the metamorphic envelope, in the border zone of the common Tatra granite (Gawęda et al. 2005) and as enclaves inside the High Tatra (Poller et al. 2001a). High Tatra granite (I/S-type) occurs in the eastern part of the massif (Fig. 1b). It is characterised by the abundance of mafic enclaves and xenoliths of country rocks (Gawęda 2009). The age of that granite body is still problematic: accessible zircon U-Pb data point out ages from 345–335 Ma (Gawęda 2008; Burda 2010) to 314 Ma (Poller et al. 2000). Goryczkowa-type granites (S-type), characterised by oriented fabric, were distinguished in the northern part of the crystalline core (Fig. 1b) and dated at 356 Ma (Burda and Klötzli 2007).

Fig. 1
figure 1

The geology of the Tatra Mountains. a simplified geological sketch of the Carpathian chain. Explanations: 1—Carpathian Foredeep; 2—Outer Carpathians; 3—Pieniny Klippen Belt; 4—Central Western Carpathians; 5—Dacides and South-East Carpathians; 6—Neogene volcanics. b geological map of the Tatra Mts block (after Kohút and Janak 1994; Bac-Moszaszwili 1996; Gawęda et al. 2005)

Full size image

Sampling and analytical techniques

Rock samples weighting about 0.5 to 5 kg were collected from the slopes of Starorobociański Mt, near the contact zone of the common Tatra type granitoid body with the metamorphic country rocks and quartz diorite sill (Fig. 1b). In the investigated rock-sequence three textural varieties of rocks were distinguished: 1—equigranular, relatively mafic (M’ = 35–28; mafic index M’ = Bt + Amph + Ttn + Ep + Opq) biotite and/or amphibole tonalites, fine- to medium-grained; 2—ocellar titanite-rich tonalites and diorites; 3—porphyritic granodiorites-monzogranites with zoned alkali feldspar and plagioclase porphyrocrysts, representing the most felsic variety of the rocks in question.

Mineral analyses were carried out in the Inter-Institution Laboratory of Microanalyses of Minerals and Synthetic Substances, Warsaw (CAMECA SX-100 electron microprobe; 15 kV, 20 nA).

Whole-rock samples were analysed by ICP-ES for major and LILE trace elements and by ICP-MS for HFSE and REE in the ACME Analytical Laboratories, Vancouver, Canada, using sets of internationally recognized standards, according to procedures described on http://acmelab.com. REEs are normalized to C1 chondrite (Sun and McDonough 1989).

Zircon crystals were separated using standard techniques (crushing, hydrofracturing, washing, Wilfley shaking table, Frantz magnetic separator and handpicking). The separation was carried out in the Institute of Geological Sciences, Polish Academy of Sciences, Cracow. Zircon grains were selected for morphological study using scanning electron microscopy and then imaged by panchromatic cathodoluminescence using a FET Philips 30 electron microscope (15 kV and 1 nA) at the Faculty of Earth Sciences, University of Silesia, Sosnowiec, Poland.

Zircon crystals were analysed in the Geochronology Laboratory, Institute of Geology at the University of Vienna. Zircon 206Pb/238U and 207Pb/206Pb ages were determined using a 193 nm solid state Nd-YAG laser (NewWave UP193-SS) coupled to a multi-collector ICP-MS (Nu Instruments HR). Ablation in a He atmosphere was either spot- or raster-wise according to the zircon CL zonation patterns. Spot analyses were 15–25 μm in diameter whereas rastering line widths were 10–15 μm with a rastering speed of 5 μm/sec. The calculated 206Pb/238U and 207Pb/206Pb intercept values were corrected for mass discrimination from analyses of standards 91500 (Wiedenbeck et al. 1995) and Plesovice (Slàma et al. 2006) measured during the analytical session. The correction utilizes regression of standard measurements by a quadratic function. A common Pb correction was applied to the final data using the apparent 207Pb/206Pb age and the Stacey and Kramers (1975) Pb evolution model. Final U/Pb ages were calculated with 2σ errors using the Isoplot/Ex program, version 3.00 (Ludwig 2003). Details of analytical procedures and data reduction schemes are given in Klötzli et al. 2009).

Mineral abbreviations used here follow these proposed by Whitney and Evans (2010).

Textures description

Plagioclase-sphene ocellar texture

Plagioclase-sphene ocelli can easily be recognized in diorite-tonalite rocks as lensoidal and spherical light patches up to 0.8 cm in diameter (Fig. 2a, b). Two varieties of this texture were recognised: mafic, with M’ at 29 and felsic with M’ at 18. Plagioclase-sphene ovoid ocelli are composed of sphene crystals constituting about 18–20 vol.% of the ocelli, in respect to the type of hybrid, usually intergrown with quartz and/or plagioclase crystals, sporadically with apatite inclusions, mantled by plagioclase + quartz + K-feldspar aggregates. Locally, plagioclase crystals forming the ocelli are idiomorphic and show normal zonation with andesine cores (An36-An31) and oligoclase mantles/rims (An21-An28). K-feldspars are xenomorphic and show reverse Ba-zonation (Cn2-4 in the cores and Cn5-9 at rims; Fig. 3a). Sphene crystals, forming the core of the ocelli (Fig. 2b), show patchy zoning (Fig. 3b), due to Al, Fe, Y and Ce substitution in the Ti and Ca sites (Table 1). Locally in the central, sphene-rich part of ocelli, allanite-epidote crystals with sector zoning can be found (Table 2; Fig. 3c).

Fig. 2
figure 2

Textures of hybrid rocks, suggesting mixing-mingling: a polished hand-specimen of hybrid rock, showing the presence of numerous plagioclase-sphene ocelli (light patches); b pseudo-mapping of a fragment of plagioclase-sphene ocellus, with idiomorphic plagioclase crystals (blue), intergrown with sphene (yellow-orange); c microphotograph of quartz-horblende-biotite ocellus in hybrid diorite. Note the presence of elongated mafic clots, crossed nicols; d microphotograph of zoned plagioclase crystal, showing calcic spikes, underlined by sericitization; white circles—anorthite content in plagioclase, crossed nicols; e microphotograph of a part of K-feldspar phenocryst with inclusion zones, underlying the chemical zonation (see Fig. 4a), crossed nicols; f microphotograph of two types of apatite: prismatic with dusty core, broken and seal by biotite, and acicular one, forming inclusions in plagioclase, crossed nicols; g microphotograph of the contact zone of diorite (left) with granite (right) with quartz crystal, poikilitically interlaced with biotite and hornblende, crossed nicols; h micrographic texture, composed of quartz and plagioclase, sealing the mafic clots, in the border zone of diorite hybrid, nicols parallel

Full size image
Fig. 3
figure 3

Back-scattered electron (BSE) images from ocellar hybrid rocks: a fine grained matrix, composed of zoned plagioclase crystals (Pl) with marked measured anorthite concentration (Anx,y = Ca/[Ca + Na + K] * 100%), zoned K-feldspar (Kfs), with marked celsian content (Cnx.y = Ba/[K + Na + Ba + Ca] * 100%) and biotite (Bt); b titanite crystals showing patchy zoning due to different Al, Fe, Y and Ce substitution (see Table 1); c allanite crystal (Aln) from the sphene-rich part of the ocellus, showing the sector zoning to epidote (Ep). It is partly overgrown by sphene (Ttn) and in contact with K-feldspar (Kfs), showing inverted barium zonation (marked celsian content)

Full size image
Table 1 Selected microanalyses of zoned titanite crystals and their crystal-chemical formulae (20 O2-)
Full size table
Table 2 Representative microanalyses and crystal-chemical formulae (25 O2-) of allanite-epidote minerals
Full size table

The ocelli are surrounded by the fine- to medium-grained matrix, composed of plagioclase, biotite and quartz, with accessory K-feldspar, apatite, zircon and opaque minerals, classified as biotite-hornblende quartz-diorite. Two types of biotite crystals could be distinguished here: blade-shape “hydrogenic” biotite (sensu Hibbard 1991), characterised by Ti contents of about 0.37–0.39 a.p.f.u. (atoms per formula unit), found in the vicinity of ocelli, and “platy” biotites, present elsewhere in the matrix, showing Ti content of 0.31–0.34 a.p.f.u. (Table 3). Plagioclase, found in the matrix, is generally oligoclase-andesine in composition (An21-37) but the zonation patterns are complicated by the calcic spikes (An41-An50). Remnants of hornblende, partly replaced by chlorite, are classified as ferro-edenitic hornblende to ferro-hornblende in the cores with actinolitic hornblende or actinolite at rims (Table 4).

Table 3 Representative microanalyses of biotite from hybrid granitoids (G) and diorites (D) and their crystal-chemical formulae (22 O2-)
Full size table
Table 4 Microprobe analyses and crystal-chemical formulae of amphiboles (13-CNK) from hybrid rocks
Full size table

Hornblende and biotite rimmed quartz ocelli

These are present in medium-grained, hornblende-bearing granite varieties. The ocelli consist of a quartz core rimmed by ferro-edenitic hornblende (Table 4) and/or biotite. The diameter of the ocelli reach 2.5 cm in length (Fig. 2c).

Plagioclase crystals with spike zoning

In all investigated rocks the presence of plagioclase crystals was noted, mostly of oligoclase composition, showing the sharp compositional discontinuities (calcic spikes) in the range An40-50 (Fig. 2d). This texture was described originally by Wiebe (1968) and then found in many magmatic rocks (i.e. Hibbard 1991, Baxter and Feely 2002) where it was attributed to magma mixing process. The exception is the matrix surrounding the titanite ocelli, where plagioclase crystals are generally more calcic (An30-50), showing normal zonation.

K-feldspar phenocrysts with inclusion zones

Coarse-grained porphyritic granite contains both normal and reversely zoned K-feldspar phenocrysts, sporadically mantled by albite, sometimes with resorption features at the border (Fig. 2e). In the K-feldspar phenocrysts the zones of inclusions are abundant, underlying the chemical zonation. Normally zoned phenocrysts have 3.7–5.2 at% of celsian component in the core and 1.5–2.0 at% in the mantle, with a local increase up to 3.1 at% celsian at the rim (Fig. 4a). The celsian contents in the reversely zoned phenocrysts change from 0.7–1.35 at% in the core, 0.5–1.1 at% in the mantle (with local peaks at 1.5–1.8, following the plagioclase rows, attached by synneusis (sensu Vernon 2004) to 2.3–2.6 at% near the margin (Fig. 4b). Such changes in Ba content could be understand in terms of mixing of more felsic, Ba-poor magma with mafic, Ba-enriched, more oxidized magma (Long and Luth 1986; Słaby and Galbarczyk-Gąsiorowska 2002; Słaby and Götze 2004).

Fig. 4
figure 4

Two examples of celsian (Cn) and albite (Ab) components distribution in alkali feldspar phenocrysts from hybrid rocks: a K-feldspar showing the normal barium zoning, expressed as celsian content (at% Cn), underlined by plagioclase rows, attached by synneusis, expressed as peaks of albite (Ab) content; b K-feldspar with inverted barium zoning, with no inclusions of plagioclase. Celsian atomic percent (at% Cn) calculated as Ba/(K + Ba + Na + Ca) * 100% from alkali feldspar formula

Full size image

Mafic clots

The border zone at the contact of quartz-diorite with granite is composed of mafic aggregates (“clots”), 2–6 mm in diameter, included in the felsic matrix. The matrix represents the quenched melt, and shows a micrographic texture (Fig. 2h) composed of quartz and plagioclase (An25-29), locally antiperthitic. The mineral components of mafic clots are: biotite, hornblende, opaque minerals (ilmenite-magnetite exsolution), allanite, and locally pseudomorphs after pyroxene.

Mixed apatite morphologies

Two generations of apatite 1) prismatic, with dusty cores, and 2) acicular (Fig. 2f) could be found in all the lithological types of magmatic rocks, including the plagioclase-sphene ocelli. Prismatic, zoned apatites (sometimes broken; Fig. 2f) represent the earlier crystallization episode from the fractionated magma. The acicular apatite crystals crystallized as a result of mafic magma quenching (especially when enriched in phosphorus). The presence of acicular apatite alone was thought to be one of the textures typical of magma mixing (Hibbard 1991), but the coexistence of two different apatite morphologies might be also explained by magma mixing—mingling process (Baxter and Feely 2002).

Poikilitic plagioclase and quartz crystals

The albitic plagioclase and quartz crystals, full of biotite, hornblende, and K-feldspar inclusions, are present in all lithologies in the mixing-mingling zone, but they are especially abundant at the contact of granite with diorite and inside the quenching zone (Fig. 2g).

Hybrid rocks chemistry

The hybrid rocks are peraluminous in composition (ASI > 1.1; Fig. 5a; Tables 5 and 6) subalkaline (Fig. 5b) and intermediate in Mg-number (#mg = 0.42–0.78; Tables 5 and 6). They plot in the high-K calc-alkaline field of the K2O versus SiO2 diagram (Fig. 5c). Generally hybrids show variable predominance of sodium over potassium (Na2O/K2O = 1.0–1.9). More felsic lithologies sporadically show K2O enrichment (Na2O/K2O = 0.6–0.9). Following the major elements, Rb/Sr ratios are low to moderate (0.08–0. 68, Fig. 5d; Tables 5 and 6). On the triangular Q-Ab-Or diagram analysed samples of hybrid granitoids plot mostly close to quartz-feldspars cotectics for pressures between 2–5 kbar, while ocellar hybrids plot near the cotectic for 10 kbar, forming a linear trend towards the quartz-diorites (Fig. 6). The hybrid quartz-diorites (mafic precursors) are metaluminous in composition, plot in the lower part of sub-alkaline field, and are characterised by extremely low Rb/Sr ratios (Figs. 5a, b, d; Table 5).

Fig. 5
figure 5

Variation diagrams for ASI a and major elements b, c, e–i versus SiO2, Rb/Sr versus SiO2 (d) and Nd versus Th (j) of the hybrid granitoids (circles), ocellar hybrids (squares) and quartz-diorites (triangles)

Full size image
Table 5 Chemical composition and selected petrological indicators of hybrid granitoids from the Western Tatra Mountains
Full size table
Table 6 Chemical composition of selected hybrid granitoid and quartz-diorite rocks from the Western Tatra Mountains
Full size table
Fig. 6
figure 6

Plot of the hybrid rocks composition in the Qtz-Ab-Or compositional space. Stairs—the melting minima for 1, 2, 5 and 10 kbar of pressure and water under-saturated conditions (after Holtz and Johannes 1991); circles—hybrid granitoids; squares—hybrids showing ocellar texture; triangles—hybrid quartz-diorites (after Gawęda et al. 2005)

Full size image

On variation diagrams TiO2, Fe2O3 and CaO hybrid granitoids and quartz diorites form one linear trend (Figs. 5e, g, h), while ASI, Al2O3 and P2O5 of these rocks form two different trends, crossing in the area of samples rich in titanite-feldspar ocelli (Figs. 5a, f, i).

Chondrite (C1)-normalized REE patterns show either negative or almost no Eu anomalies (Eu/Eu* = 0.62–1.05), except for the porphyritic granite with a positive Eu anomaly (Eu/Eu* = 1.24). The REE fractionation is very variable: from low (CeN/YbN = 4.5–10) to high (27.3–55.6; Fig. 7a, b; Table 5 and 6). The quartz diorites present in the same area (Gawęda et al. 2005) are characterised by weakly negative or lack of Eu anomaly (Eu/Eu* = 0.777–1.065) and moderate REE fractionation (CeN/YbN = 12.72–18.04; Table 5; Fig. 7a, b–grey shadow area for comparison).

Fig. 7
figure 7

Rare earth elements patterns, normalized to C1 chondrite (after Sun and McDonough 1989), in hybrid granites a and in ocellar rocks b. The shaded area shows the range of REE patterns of quartz-diorites from the Tatra Mountains (after Gawęda et al. 2005)

Full size image

Temperature and pressure calculations

The application of Schmidt (1992) hornblende geobarometry calibration (Table 7) gives an apparent pressure range of 6.1 kbar to 4.6 kbar, in accordance with earlier results obtained for quartz diorites from the same area (Gawęda et al. 2005). Temperature of crystallization was calculated using the Blundy and Holland (1990) geothermometric calibration and comparatively, the Vyhnal et al. (1991) calibration as well as the similar calibration proposed by Gawęda (2009). Ternary feldspars geothermometry (Fuhrman and Lindsley 1988; Nekvasil and Burnham 1987) was applied, giving the disequilibrium temperatures, although plotting in the narrow range of 695°C to 738°C for assumed 4–6 kbar of pressure (Table 7). Anti-perthites, found in the micrographic texture were also used to calculate the exsolution equilibrium temperatures, using ternary feldspars geothermometry, and point out the range of 551-568°C. The whole rock Zr contents were used to calculate the magma temperature (Harrison and Watson 1983), which plot in the bracket of 810°C–770°C.

Table 7 Temperature and pressure calculations
Full size table

Zircon characteristics and geochronology

Zircon crystals from the hybrid granitoid showing ocellar textures (sample SP2; Table  6) were selected for dating. Crystals are clear and colourless to pale yellow. Zircons are typically 200–300 μm in length, euhedral, short to long prismatic, with aspect ratios of 1:2 to 1:5 (Fig. 8). There are two morphological types of crystals. Type 1 shows a predominance of [110] prism forms and the presence of two pyramids [101, 211] with the latter more dominant (subtypes S2, S6, S7 in Pupin’s typological classifications; Fig. 9). This morphological type is characteristic for zircon crystals from S-type aluminous granitoids (Pupin 1980). Type 2 zircons have a well-developed [110] prism and [101] pyramid forms (subtypes S5, L5, L4; Fig. 9). According to the genetic classification (Pupin 1980) this type is typical for zircon from sub-alkaline and alkaline series granitoids of crustal plus mantle or mainly mantle origin.

Fig. 8
figure 8

Secondary electron (SE) images of zircon crystals from hybrid granitoid showing ocellar textures (sample SP2). Descriptions of zircons are included in the text

Full size image
Fig. 9
figure 9

Distribution of morphological types of zircon from hybrid granitoid with ocellar textures (sample SP2) on Pupin’s (1980) diagram

Full size image

Some zircon grains revealed intermittent dissolution surfaces between well-developed oscillatory zoning (Fig. 10). These might fingerprint the corrosion or resorption events during evolution of zircon crystals (e.g. Vavra 1990, Köksal et al. 2008). Some grains exhibit homogeneous to weakly growth-zoned cores with bright luminescence surrounded by euhedral overgrowths with oscillatory zoning. Zircon cores are sub-rounded and the contacts with surrounding oscillatory-zoned rims are irregular. Locally, recrystallization patches (loss of oscillatory zoning) are also present (Fig. 10).

Fig. 10
figure 10

Cathodoluminescence (CL) images of zircon crystals from hybrid granitoid with ocellar textures (sample SP2). Grains no. 5, 6, 7, 8 and 9 show magmatic oscillatory zoning from centre to margin. Grains no. 1 and 10 contain an older core surrounded by a younger rim with oscillatory zoning. The white lines show the approximate locations of laser ablation trenches and are not to scale. Analyses from the oscillatory-zoned zircon zones yield a concordia age of 368 ± 8 Ma. Two data points from inherited cores give a concordia age of ca. 530 Ma (Fig. 11; Table 9)

Full size image

Microprobe analyses were performed on zircon sections polished parallel to the c axis. In analysed grains oscillatory zoning of HfO2 is visible (Table 8). A positive correlation between P and Y, suggesting the xenotime-type substitution Zr4++Si4+↔(Y, REE)3++P5+ (Speer 1982), is observed. The crystal domains with oscillatory zoning are characterized by Zr/Hf ratios in the range of 35–52 and Th/U > 0.1 (Table 8).

Table 8 Microprobe analyses of selected zircon crystals from hybrid granitoids showing ocellar textures (sample SP2 (in wt% oxides)
Full size table

Eleven LA-MC-ICP-MS measurements on seven zircon crystals from the same rock (SP2) were made (Table 9). All data points are concordant within the assigned error (Fig. 11). Nine analyses from the oscillatory-zoned zircon zones yield a concordia age of 368 ± 8 Ma (MSWD = 1.1). Two data points from inherited cores give a concordia age of ca. 530 Ma.

Table 9 LA-MC-ICP-MS U-Pb zircon data from hybrid granitoid with ocellar textures (sample SP2)
Full size table
Fig. 11
figure 11

Concordia plots of LA-MC-ICP-MS U-Pb zircon analytical results from the oscillatory-zoned zircon zones from hybrid granitoids showing ocellar textures (sample SP2)

Full size image

Discussion

The Western Tatra granitoids were previously interpreted as typical S-type granites (Poller et al. 2001a). The textures found in the analysed rocks advocate that mixing and mingling of magmas of different chemistry occurred during the granitoid magma emplacement. The presence of mafic clots and associated micrographic intergrowths at the border zone between the mafic and felsic portions suggest the clots have originated in mafic magma and cooled relatively quickly. Quenching is thought to be a process responsible for the acicular apatite formation, both in the matrix and as the inclusions in feldspars (Hibbard 1991; Baxter and Feely 2002). Anorthite spikes in plagioclases are also interpreted in terms of felsic and mafic magma mixing, as well as titanite-feldspar ocelli. The zoned allanite-epidote crystals (Fig. 3b) are typical of mafic, oxidised magmas (Broska and Uher 1991). The high content of Ba in fine-grained matrix K-feldspar crystal and their inverted chemical zonation might be also attributed to magma mixing (Long and Luth 1986; Słaby and Götze 2004). Hornblende and biotite rimmed quartz ocelli are explained as the result of magma mixing, taking place during the crystal nucleation (Hibbard 1991; Baxter and Feely 2002), i.e. rather early in the granitoid magma history.

The internally zoned alkali feldspar megacrysts, with inclusion zones, resorption and sporadically discontinuous albite mantling might be a result of both sub-isothermal decompression (Nekvasil 1991, Eklund and Shebanov 1999) and mafic and felsic magma mixing (e.g. Hibbard 1991, Słaby and Götze 2004).

The linear trends observed in variation diagrams for hybrid granites and quartz-diorites (Figs. 5a-i) indicate mixing as a predominant factor governing the evolution of the magma. Although the quartz-diorites are hybrid rocks, fractionation was also shown to have occurred in their parent magma (Gawęda et al. 2005), in contrast to the hybrid granites where no fractionation trend was noted.

Nd and Th are thought to be elements representing contrasting magmatic sources: Nd enrichment is typical of mantle derived magmas while Th is enriched in those derived from the upper crust. The Nd versus Th brings the important information about the hybrid rocks position (Fig. 5j). Most of hybrid granitoids and some diorites form one trend with Nd/Th = 2–4, suggesting an interplay between monazite-bearing crustal rocks and mantle-derived magma (Bea et al. 1999). Two quartz-diorite samples and ocellar titanite-rich hybrids are Nd-enriched (Nd/Th = 4–9), suggesting predominance of mantle component, while the two other hybrids plot as Th-rich (Nd/Th = 1.9-2.5), representing the crustal-dominated component (Table 5 and 6; Fig. 5j). Those observations are consistent with the suggestions of Poller et al. (2000), based on the Sr, Nd and Pb isotopic composition, pointing out the slab break-off model and mantle plume for the generation of diorite magma.

Relative to the quartz diorites the hybrid granitoids show lower total REE content and LREE enrichment (expressed as CeN/YbN). The negative Eu anomaly is typical of granites crystallizing under reduced conditions. K-feldspars are common carriers of Eu, causing the positive Eu anomaly in one porphyritic granitoid. The ocellar titanite-rich hybrids show chondrite-normalized REE patterns transitional between granitoids and diorites, with Eu/Eu* ~ 1, which suggests more oxidizing conditions and a mantle influence (Table 5 and 6; Fig. 6).

Based on zircon typology, these granitoid samples are classified as hybrid calc-alkaline series granitoids (Group-4 of Pupin 1980; Fig. 9). CL images of the investigated zircons reveal domination of euhedral crystals with fine to medium-scale oscillatory zoning. In the early stages of growth [121] pyramids predominate, typical of crystals from granitoids with dominantly S-type characteristics. Then, the morphology of crystals changes, favouring [101] pyramids, common in hybrid-type granitoids (e.g. Pupin 1980; Schermaier et al. 1992; Köksal et al. 2008). The occurrence of oscillatory zoning of Hf in zircons suggests heterogeneities in the crystallization environment, or repeated fluctuations of the Hf concentration in the magma during zircon growth (e.g. Benisek and Finger 1993). This might be a result of rapid change in chemical conditions at the latest stages of crystallization, caused by the influx of new magma and consequently, may be related to magma mixing/mingling processes. The presence of intermittent dissolution surfaces between well-developed oscillatory zoning (Fig. 10) also supports this suggestion.

An overall enrichment in P and Y from core to rim is observed in analysed zircons (Table 8). The xenotime substitution in zircon is obviously constrained by the availability of Y and P in the magma. Therefore, the crystallization of apatite of mixed morphologies with respect to zircon (in the absence of monazite) is the key factor. Only if some P is still available after the complete crystallization of P-bearing minerals (apatite), it can enter into zircon in significant amounts (Caironi et al. 2000).

U-Pb isotopic data obtained by LA-MC- ICP-MS analyses of oscillatory growth-zones (Fig. 10) yield a concordia age of 368 ± 8 Ma (2 sigma; Fig. 11). This age is interpreted as reflecting the time of zircon formation during a magmatic event.

Only a few crystals have subhedral cores. These are interpreted as xenocrysts, probably inherited from the crustal source-rocks of the magma. The zircon cores yield an age of ca. 530 Ma which reflect their crystallization age. A similar inherited zircon age signature was found in paragneisses (Kohút et al. 2008) and in orthogneisses from the crystalline basement of the Tatra Mountains (Burda and Klötzli 2010) as well as from another crystalline complexes from the Central Western Carpathians (e.g. Putiš et al. 2008; Putiš et al. 2009). The presence of these oldest meta-magmatic rocks in the Western Tatra Mts. could be linked to the fragmentation of the northern margin of Gondwana (Burda and Klötzli 2010).

The mixing-mingling of felsic and mafic magmas, predating the main granite intrusion and influencing the development of the Western Tatra granite, sheds new light on the origin of this granitoid body. The Western Tatra granite, previously interpreted as purely S-type, is in fact a composite pluton, showing common features with the I/S High Tatra granite portion. The age of the hybrid rocks from the Western Tatra Mountains marks subduction processes at the Laurussia margin followed by the collision of microterranes, and finally of the Gondwana promontory.

Conclusions

  1. 1.

    Granitoids from the Western Tatra Mts., interpreted previously as the products of crystallization from purely S-type melts, represent in fact hybrid melts.

  2. 2.

    LA-MC-ICP-MS U-Pb zircon age calculations (368 ± 8 Ma) indicate an Early Variscan magmatic episode. As the 368 Ma age is found in zircon rims it could be interpreted as the oldest Variscan magmatic event in that part of the Tatra Mountains.

  3. 3.

    Textural features, together with mineralogy and chemistry, point out the presence of mixing/mingling phenomena, which started from early stages of magma evolution. The vast distribution of textures generated by mingling-mixing processes, together with the rocks chemistry, suggests that these processes governed the formation of the Western Tatra granitoid pluton.

  4. 4.

    The hybridization event defined here, which predates the main magmatic activity in the Tatra Mountains, provides new insight into the paleotectonic reconstructions of the Early Variscan realm in the Western Carpathians.